|Publication number||US4723311 A|
|Application number||US 06/807,209|
|Publication date||Feb 2, 1988|
|Filing date||Dec 11, 1985|
|Priority date||Dec 14, 1982|
|Also published as||DE3246301A1, EP0111332A2, EP0111332A3, EP0111332B1|
|Publication number||06807209, 807209, US 4723311 A, US 4723311A, US-A-4723311, US4723311 A, US4723311A|
|Inventors||Steven Moustakas, Hans-Hermann Witte|
|Original Assignee||Siemens Aktiengesellschaft|
|Export Citation||BiBTeX, EndNote, RefMan|
|Patent Citations (8), Referenced by (28), Classifications (13), Legal Events (3)|
|External Links: USPTO, USPTO Assignment, Espacenet|
This is a continuation of application Ser. No. 560,350 filed Dec. 12, 1983 now abandoned.
The present invention relates to a method for detecting data collisions in an optical data bus.
Circuits which implement a method for the above purpose have been proposed in the older German patent application P No. 32 24 664.1 (VPA 82 P 1509). A U.S. patent application for the invention was filed on June 16, 1963, and was assigned Ser. No. 505,432. Its contents are herein incorporated by reference. The proposed methods for collision recognition are based on a level evaluation of the pulses arriving in a receiver or on a bit by bit comparison of the pulses sent and received.
It is an object of the present invention to provide additional advantageous methods for the detection of data collision in an optical bus.
It is a further object of the present invention to disclose circuit embodiments for the detection and recognition of data collisions.
According to a preferred embodiment of the invention collisions are detected directly by all subscribers to the optical bus via two analog comparators. Through the analog comparators an amplitude evaluation is effected for the determination of the superposition of two or more data sequences simultaneously on the optical data bus. The present implementation is particularly effective if received optical power fluctuations at the site of any receivers which are present in the various subscribers to the bus do not exceed 3dB. According to the present embodiment JAM signals are not generated to indicate to others subscribers the occurrence of a data collision.
In accordance with the a further embodiment of the invention, to begin with one or more transmitting subscribers detect collisions. According to the present embodiment power fluctuations at the site of any receiver do not limit the functionality of the circuit. A collision which is detectable in a transmitting subscriber is communicated to all other subscribers by a JAM signal. The JAM signal comprises a bit pattern which is unique and different from the typical data signals.
A still further preferred embodiment of the present invention discloses a method for detecting collisions in which, during the time which corresponds to the loop-time or turn-around time of the system bus, each subscriber interrogates whether its receiver is receiving data signals from the data bus. This method is referred to as "collision recognition operating in accordance with the monitoring principle".
According to the first embodiment above in which comparators are used, collisions are recognized directly by all subscribers. This comprises an especially simple method. However, it is realizable only if the maximum received optical power difference at any one receiver does not exceed 3 dB.
According to the second embodiment above restriction as to power fluctuations is not important. However, collisions can be detected only by subscribers transmitting at a given time, which in turn then communicate this state i.e. that a collision has occured to all non-transmitting subscribers by the so-called JAM signal. This signal has a distinctive pattern which can be differentiated from the normal useful data bits.
The two methods above provide the basis for the realization of asynchronously operating optical data buses in which bus access, collision and interrupt requests can be recognized in all subscribers.
According to the third mentioned embodiment which operates according to the monitoring principle, a simple control circuit for the recognition of data collisions and bus access status in optical buses is available. The method may be used in particular in star buses with asynchronous access procedures which purposely does not impose power equidistribution requirements and in which the position of various subscribers along the data bus can be easily accounted for. Signal levels or signal powers according to this last mentioned method are less critical for the detection of data collision. Also, the last method may be implementated by a greatly simplified circuit and critical bottlenecks in a coupler associated with the typical circuitry are eliminated.
Other features and advantages of the invention will be apparent from the following description of the preferred embodiments, and from the claims.
For a full understanding of the present invention, reference should now be made to the following detailed description of the preferred embodiments of the invention and to the accompanying drawings.
FIG. 1 shows a block diagram of a circuit for collision and free-line or data bus access recognition.
FIG. 2 shows a diagram or a circuit for a transmitter in the circuit according to FIG. 1.
FIG. 3 shows a circuit for the recognition of the free-line or data bus access and of collisions, which is based on the last cited preferred method embodiment in the SUMMARY OF THE INVENTION section.
FIG. 4 shows a receiver control circuit.
FIG. 5 illustrates schematically a bit pattern of a data packet in conjunction with a Valid signal.
For the description which follows, the following applies:
The time constant τi (i=1 to 7) of the (retriggerable) monostable multivibrator i, designated as (R.) M.M.i. for short, corresponds to the time of the unstable state of the multivibrator.
The time constants given in numbers relates to a bus length of 600 meters. For a refractive index of a glass fiber core of 1.5, the transmit time of a pulse on a glass fiber length of 600 m is 3 us.
An interrupt request is represented by a short positive pulse.
The time between the start and end of useful data is enclosed in the time period during which the "validity" signal is in a logic "1" state.
The data rate of the system is 16 Mbit/s. The scrambler needed for scrambling the data signals is designed so that at its output not more than eight "1"s or "0"s appear in uninterrupted succession.
The state "Send no data" is represented by an uninterrupted succession of bits of the binary state "0" which is longer than eight bits.
JAM signals are triggered by interruption requests or collisions and comprise about 30 bits of binary state "1" in uninterrupted succession.
The circuit arrangements as described in the U.S. patent application referenced above are based upon the following assumptions:
(1) data start and end of the data supplied by the computer are indicated by pulses on separate lines;
(2) collisions need be recognized only in the transmitting subscriber; and
(3) interrupt requests need to be recognized.
The bits of a transmitting subscriber are compared with the bits which arrive in the receiver of the same subscriber. A delay circuit is provided, such that each bit of the transmitter coincides in time with the bit coming from the same transmitter and which has returned to the receiver of the transmitting subscriber after passing through the bus length. In another circuit version, the delay element, which is to be matched in time, is replaced by an analog comparator which, together with another analog comparator always existing in the actual receiver loop, make it possible to detect collisions by evaluation of the amplitudes of the bit streams. A circuit for implementing the above has been disclosed in the aforementioned prior patent application (German patent application P No. 32 24 664.1).
In accordance with the present invention, the requirements for the collision and free-line circuit are made stricter essentially in three points. In this proposal the following three modified requirements have been taken into consideration.
1. The data start and end are to be indicated, not by two separate lines or "trigger pulses", but by one pulse, the duration of which comprises the data transmissions, i.e. the sending of the header and dummy bits plus actual or useful data. As has been mentioned before, this pulse is referred to as the "validity" signal.
2. Collisions are to be recognized in all computers, not--as before--in transmitting computers or subscribers only.
3. The system is to recognize interrupt requests.
As far as the second requirement is concerned, two implementations must be distinguished.
I. If the maximum received optical power fluctuations in any receiver does not exceed approximately 3 dB, preferably 2 dB, it is possible with the level evaluation proposed in the cited older patent application to recognize directly collisions not only in the transmitting but in all computers by way of two comparators. The thresholds in the threshold value comparators are selected so that when only one bit stream is present, only one comparator responds, and this indicates that there is no collision. But if at least two bit streams overlap, then the other comparator also responds, indicating a collision. A clear distinction between collision and no collision is possible with this principle only if the input power fluctuations at the receiver are not so great that the sum of the amplitudes (at the output of the amplifier) of two bit streams can become smaller than or equal to the amplitude of one other bit stream. This method does not require a JAM signal.
II. If the star coupler section of a bus does not fulfill the condition stipulated for power levels, collisions can be recognized only in the transmitting computers. The detection of a collision must be communicated to all computers via a JAM signal. A JAM signal must be a bit pattern which clearly differs from useful data and from the state "Send no data". Specialized circuitry is required for the generation and recognition of a JAM signal.
In the description below which pertains to an embodiment according to FIG. 1, it is assumed that a JAM signal is needed. It must be possible to produce and to recognize them. In the drawings certain functions are represented by logic block units. Thus, for example, needed power drivers, the logic function of which corresponds to an inverter, are represented partly by inverters.
FIG. 1 illustrates the interplay of the logic block units.
The circuit for recognition of free line, i.e. of the state in which the bus is free of data signals, is the same as described in the previously mentioned patent application. The time constant τ1 of the retriggerable monostable multivibrator R.M.M.1 which generates the free-line signal is chosen to be about 5 us long in the case here described. It is thereby ruled out that long uninterrupted sequences of bits of binary state `1`, as they occur, e.g. in a JAM signal, will generate a free-line signal. If, in fact, τ1 is not large enough, then for logic `1` sequences which are longer than τ1, the output Q of the multivibrator R.M.M.1 becomes `1`, because in the `1` sequence a positive edge change or respectively a positive edge cannot take place or occur.
As has been explained in detail in the older patent application in a transmitting computer in case of a collision x7 =`1`. Owing to this, x"18 =x18 =`1` (x18 corresponds to output x15 in FIG. 2 of the older patent application). Because of the monostable multivibrator M.M. 4 x19 becomes `0`. If x7 becomes `1`, x'22 =`1`. This collision signal x'22 switches x17 to `0`, so that x'11 becomes `1`, and a logic `1` remains (a measure taken in the computer). As long as x19 =`0` (τ4), the LED sends a logic `1`. This JAM signal is selected about 30 bits long. Thereafter x19 =x'11 =`1` and the LED sends only a `0`. The time constant τ6 is chosen to be a little (100 ns . . . 200 ns) greater than the time constant τ3. This is because, when the JAM signal generated by a collision recognition comes back to the place of the origin after running through the bus length, it is to be prevented that a JAM signal is again produced. The returning JAM signal may, however, produce a collision (x7 =`1`). As long as x7 =`1`, x18 must remain on `1` through a corresponding time constant τ6 in order that x19 will not again go to `0` and trigger a new JAM signal.
Upon occurence of an interrupt pulse (a computer wants to inteDrrupt the existing state and possibly transmit itself), x"18 becomes `1` and x20 becomes `1` for the time which corresponds to the time constant τ4, i.e. the LED produces a JAM signal. Simultaneously with the interrupt pulse, x17 is switched to `0` (and remains `0`). After the JAM signal the LED (light-emitting-diode) is turned off.
The circuit for recognition of the JAM signal corresponds, roughly, to a counter which recognizes that for instance more than 30 bits of binary state `1` have come out of the comparator K1 of the receiver in a continuous uninterrupted sequence. This state is uniquely correlated with the JAM signal. The time constant of the retriggerable monostable multivibrator R.M.M.5 of the JAM signal recognition circuit is chosen more than twice the time required for the longest uninterrupted flow of `1` from the scrambler. Let it be assumed, for example, that not more than eight `1`s, or `0`s can be generated by the scrambler in uninterrupted succession. The time constant τ5 of the multivibrator R.M.M.5 is then chosen for example to be 20 bits long. During "Data transmission" within this time a positive edge change then always occurs, so that data output Q remains `0`. Now if a JAM signal occurs, for instance for a sequence of 30 bits of `1`s, Q will remain `0` for a period of at most 20 bits and then change over to a `1` state. If a `1` of a data sequence gets into the multivibrator R.M.M.5, the `0` edge output at P, which is triggered by the positive edge of this `1` appears delayed by about 15 ns. Then for this time of 15 ns, x21 would be x'2 =x21 =`1`. This data would be incorrectly recognized as a JAM signal. However, if x2 is delayed by at least 15 ns, the incoming data sequence is not interpreted as a collision. Now if a JAM signal of e.g. 30 successive `1` actually occurs, Q will remain `0` for a period of at most 20 bits and then becomes a `1`. Together with x'2 =`1` x'22 becomes `1`, and therefore the JAM signal is recognized in the form of the collision signal `1`.
In the circuit illustrated in FIG. 1, non-transmitting computers can recognize collisions only through JAM signals which originate from transmitting computers in which collisions have been recognized.
It is to be stressed once more than in the example set forth in detail according to FIG. 1 a JAM signal must be produced and recognized, because collisions are detected only in transmitting computers. Further it is assumed in the circuit according to FIG. 1 that recognition of the collisions in transmitting computers occurs by a bit-by-bit comparison of their own data. Detection of collisions in transmitting computers is possible also by two comparators, as represented in the mentioned older patent application. Lastly, in the simplest case, collisions can be recognized in all computers directly, i.e. without JAM signals, if the maximum optical input power fluctuation at any receiver as a function of all transmitters does not exceed approximately 2 dB. This is carried out again with the aid of two comparators. In this case the line x6 and the retriggerable monostable multivibrator R.M.M.3 in FIG. 1 would become superfluous.
In FIGS. 2 to 4, three control circuit units are indicated, which provide and show embodiments for the circuit corresponding to FIG. 1.
The three units consist of the transmitter control illustrated in FIG. 2, the free-line and collision recognition illustrated in FIG. 3, and of the receiver control illustrated in FIG. 4.
The transmitter control contains the unit A of FIG. 1. The transmitter control sends data in the form of data packets, each of which contains, in addition to the useful data, header bits and possibly also dummy bits as well as end-marking bits. The header bits and possibly also the dummy bits precede the useful data, while the end-marking bits mark the end of the useful data. The header bits, dummy bits and useful data, which appear during the time marked from the leading edge to the trailing edge of the validity signal, are sent by the transmitter control in scambled form.
From the scrambled data packet sent by the transmitter control the receiver control recovers the useful data. The same is true of the clock and the validity signal. In case of collision it switches the validity signal off.
The free-line and collision recognition contains also a collision recognition circuit. When this free-line and collision recognition circuit is used for the realization of the circuit according to FIG. 1, it also means therefore, that the collision recognition principle is changed, because the collision recognition circuit C in FIG. 1 is replaced by the differently operating collision recognition circuit C' according to FIG. 3.
Below there will be described only the collision recognition circuit C' and the circuit simplifications resulting therefrom in the "free-line and collision recognition circuit" according to FIG. 3.
For illustration purposes actual numbers are used. It is assumed again that the data rate on the bus is 16 Mbit/s and the distance between the subscribers is about 600 meters, that the scrambler does not permit an uninterrupted succession of more than eight `0`s or `1`s in the scrambled data packet, that header bits, always starting with a logic `1` at the beginning of a data packet, are 14 bits long, and that the end of a data packet during transfer via the bus consists of a `0` followed by eleven bits of logic state `1`. The data-end pattern is produced by the transmitter control according to FIG. 2, namely by unit D. In addition to the actual data, there are supplied by the computer the validity signal (Valid signal) and the clock, both of which must be recovered in the receiver control.
The entire data packet sent by the LED, consists of header and dummy bits, actual data and the end-marking bits, and has the pattern appearing in FIG. 5. In it, the validity signal encompasses the pattern of the data packet. The validity signal itself is not transmitted.
The time constant τ1 of the retriggerable monostable multivibrator R.M.M.1 in FIG. 3 is chosen so that the free-line state is indicated six bits later, after the entire data packet has been received by all subscribers. A JAM signal consists of an uniterrupted flow of `1`s, not less than 24 bits long, corresponding to a duration of at least 1.5 us. The four inverters from the delay V2 and accordingly take into consideration the delay through the retriggerable monostable multivibrator R.M.M.5, so that together with the time constant τ5 =1.2 us it is assured that x'6 can become `1` only during the JAM signal. From the JAM recognition signal or the collision recognition signal there is formed by the monostable multivibrator M.M.7 a signal x10 suitable for the computer. Furthermore there is produced through the JAM recognition signal or through the collision recognition signal a signal x19 which in the receiver control circuit according to FIG. 4 terminates the validity signal e26 and the recovered data flow to the computer e8, opens AND gate Uz preceding the counter Z again, and resets the counter Z to `0`, ready for counting.
When a subscriber starts to transmit, a 1-bit pulse is generated in the transmission control by the positive edge of the validity signal. This causes x17 in the J-K flip-flop FF in FIG. 3 to become `1` and the AND gate U3 to be opened. While the AND gate U3 is open, the transmitting subscriber can interrogate the bus via the retriggerable monostable multivibrator R.M.M.1 as to whether or not pulses are coming in. This monitoring time is chosen shorter than, but a equal as possible to the turn-around loop time corresponding to this subscriber. If during this time a signal is received, it must originate from another subscriber, that is, a collision has taken place. At x7 a positive pulse is generated. In addition to the effects due to a JAM recognition, pulse x7 =`1` brings about that in the transmitter control a JAM signal is produced via x18. By appropriate selection of the time constant τ7 in the monostable multivibrator M.M.7 a retriggering due to the JAM signal produced by the subscriber recognizing the collison is avoided.
x17 become `1` at the start of the validity signal. x17 becomes `0` if during the bus loop time a collision signal is recognized, or at the latest after the bus loop time. The delay of the shift register R3 in FIG. 3 corresponds to the bus turn-around or loop time.
By the OR gate OD at the K input of the J-K flip-flop FF in FIG. 3 it is ensured that not more than one collision signal is produced during the bus turn-around time.
The new collision recognition circuit is simple, entirely uncritical and reliable. It does not require synchronization of the bits and sets practically no requirements for the power fluctuations.
As the first bit of data packet is always a `1`, each subscriber needs to monitor the bus for incoming bits only for the time of one bus loop, without loss of efficiency in collision detection. Also a subscriber should monitor at most for the time duration of one turn-around, because thereafter the subscriber receives back its own data.
In the procedure according to the monitoring principle each transmitting subscriber listens to the bus, from the instant of the start of its data, which coincides with the beginning of the Valid signal. But for practical and logical reasons the monitoring time will always turn out shorter than the respective bus turn-around time. This means that there is a dead time during which there is no monitoring and therefore collisions cannot be recognized. For certain applications the probability of non-recognition of collision during the dead time may become too high.
To be able to recognize collisions also during the dead time, the procedure preferably is to detect collisions during this dead time through auxiliary hardware.
Especially suitable for this purpose is the method operating according to the principle of "collision recognition on the principle of level evaluation with JAM signal generation", because this method sets practically no requirements for the dynamic level or power fluctuations at the site of a receiver.
To be able to detect a collision during the dead time with certainty, one selects the dead time either shorter than the duration of one bit or equal to one bit, and makes sure that the data packets, specifically the header bits of the data packets, begin with two consecutive bits of binary state `1`.
It is thereby assured that in case of superposition of two bit streams that a sufficient level of signal will always occur, to which a comparator indicating a collision can respond. This means also, however, that the dead time must not be made too short relative to the length of one bit. It must be long enough for said threshold value comparator to respond.
In star networks, where the distances of various subscribers from the coupler are generally different, the monitoring times of the subscribers are correspondingly different. In any event, at least one subscriber will detect a collision and produce the respective JAM signal if the interrogation time is identical with the bus turn-around time. In this instance it is no longer necessary to distinguish between transmission faults and collisions.
In a ring bus, all collisions are recognized if one selected the monitoring time between one half and a full turn-around or loop time.
The circuits shown in FIGS. 2, 3 and 4 can be constructed with components of Texas Instruments as documented for example in "The TTL Data Book", Texas Instruments, fourth European Edition 1980. The designations LS 221, LS 123, S 112, LS 164, LS 21, LS 08, S 08, S 32, LS 04, S 04, LS 00, LS 02, LS 3 given in the FIGS. signify the designations of the components in the data book.
There has thus been shown and described a novel apparatus and method for detecting data collisions in a bus which fulfills all the objects and advantages sought therefor. Many changes, modifications, variations and other uses and applications of the subject invention will, however, become apparent to those skilled in the art after considering the specification and the accompanying drawings which disclose preferred embodiments thereof. All such changes, modifications, variations and other uses and applications which do not depart from the spirit and scope of the invention are deemed to be covered by the invention which is limited only by the claims which follow.
|Cited Patent||Filing date||Publication date||Applicant||Title|
|US3634869 *||Dec 29, 1970||Jan 11, 1972||Chia Ying Hsueh||Interpulse time interval detection circuit|
|US4114106 *||Dec 28, 1976||Sep 12, 1978||Telefonaktiebolaget L M Ericsson||Arrangement to indicate signals having a length exceeding a limit value|
|US4422179 *||Jul 15, 1981||Dec 20, 1983||Bell Telephone Laboratories, Incorporated||Electrical-optical interface network|
|US4439856 *||Feb 24, 1982||Mar 27, 1984||General Electric Company||Bimodal bus accessing system|
|US4476467 *||Jun 8, 1982||Oct 9, 1984||Cromemco Inc.||Random entry intercomputer network with collision prevention|
|US4499576 *||Aug 13, 1982||Feb 12, 1985||At&T Bell Laboratories||Multiplexed first-in, first-out queues|
|US4531238 *||Dec 3, 1981||Jul 23, 1985||Xerox Corporation||Statistical contention control for star configured communication networks|
|GB2111337A *||Title not available|
|Citing Patent||Filing date||Publication date||Applicant||Title|
|US4809264 *||Aug 7, 1987||Feb 28, 1989||Chipcom Corporation||Collision signal detection system|
|US4850045 *||Nov 2, 1987||Jul 18, 1989||Amp Incorporated||Method and apparatus for testing fiber optic hubs|
|US4943979 *||May 12, 1988||Jul 24, 1990||Farallon Computing Corporation||Local access network signal regnerator|
|US5113438 *||Jun 25, 1990||May 12, 1992||Cablesoft, Inc.||Method and apparatus for jamming infrared remote controls|
|US5119398 *||Oct 19, 1989||Jun 2, 1992||Farallon Computing, Inc.||Signal regenerator for two-wire local area network|
|US5390039 *||May 16, 1994||Feb 14, 1995||Canon Kabushiki Kaisha||Optical network for detecting collisions based on the reflection of a transmitting signal|
|US5553247 *||Dec 12, 1994||Sep 3, 1996||Alcatel Cit||Method for unblocking a multibus multiprocessor system|
|US5568476 *||Oct 26, 1994||Oct 22, 1996||3Com Corporation||Method and apparatus for avoiding packet loss on a CSMA/CD-type local area network using receive-sense-based jam signal|
|US5857075 *||Jan 11, 1995||Jan 5, 1999||Sony Corporation||Method and integrated circuit for high-bandwidth network server interfacing to a local area network|
|US5884040 *||Oct 8, 1997||Mar 16, 1999||Sony Corporation||Per-packet jamming in a multi-port bridge for a local area network|
|US5940597 *||Oct 8, 1997||Aug 17, 1999||Sony Corporation||Method and apparatus for periodically updating entries in a content addressable memory|
|US6012099 *||Dec 19, 1997||Jan 4, 2000||Sony Corporation||Method and integrated circuit for high-bandwidth network server interfacing to a local area network|
|US6122667 *||Dec 19, 1997||Sep 19, 2000||Sony Corporation||Method and integrated circuit for high-bandwidth network server interfacing to a local area network using CSMA/CD|
|US6157951 *||Apr 22, 1998||Dec 5, 2000||Sony Corporation||Dual priority chains for data-communication ports in a multi-port bridge for a local area network|
|US6256313||Mar 30, 1998||Jul 3, 2001||Sony Corporation||Triplet architecture in a multi-port bridge for a local area network|
|US6301256||Apr 22, 1998||Oct 9, 2001||Sony Corporation||Selection technique for preventing a source port from becoming a destination port in a multi-port bridge for a local area network|
|US6308218||Jun 9, 1998||Oct 23, 2001||Sony Corporation||Address look-up mechanism in a multi-port bridge for a local area network|
|US6363067||Mar 19, 1999||Mar 26, 2002||Sony Corporation||Staged partitioned communication bus for a multi-port bridge for a local area network|
|US6442168||Feb 18, 1998||Aug 27, 2002||Sony Corporation||High speed bus structure in a multi-port bridge for a local area network|
|US6617879||Mar 19, 1999||Sep 9, 2003||Sony Corporation||Transparently partitioned communication bus for multi-port bridge for a local area network|
|US6738384||Mar 19, 1999||May 18, 2004||Sony Corporation||Technique for optimizing cut-through for broadcast and multi-cast packets in a multi-port bridge for a local area network|
|US6744728||Mar 19, 1999||Jun 1, 2004||Sony Corporation & Sony Electronics, Inc.||Data pipeline timing optimization technique in a multi-port bridge for a local area network|
|US6751225||Mar 19, 1999||Jun 15, 2004||Sony Corporation||Port within a multi-port bridge including a buffer for storing routing information for data packets received in the port|
|US6816490||Mar 19, 1999||Nov 9, 2004||Sony Corporation||Statistical learning technique in a multi-port bridge for a local area network|
|US6822970||Jan 31, 2000||Nov 23, 2004||Owens-Brockway Glass Container Inc.||Glassware forming system with star network communication configuration|
|US8750903||Mar 28, 2012||Jun 10, 2014||CellAntenna Corporation||Cell phone control and localization for restricted facilities|
|US20040223509 *||Jun 9, 2004||Nov 11, 2004||Redd Matthew D.||Glassware forming system with star network communication configuration|
|WO1996013957A1 *||Oct 5, 1995||May 9, 1996||3Com Corp||Method of jamming to avoid packet loss in csma/cd-lan|
|U.S. Classification||398/36, 385/24, 370/445, 398/39|
|International Classification||H04B10/27, H04B10/278, H04L12/44|
|Cooperative Classification||H04B10/278, H04B10/27, H04L12/44|
|European Classification||H04B10/27, H04B10/278, H04L12/44|
|Sep 3, 1991||REMI||Maintenance fee reminder mailed|
|Feb 2, 1992||LAPS||Lapse for failure to pay maintenance fees|
|Apr 7, 1992||FP||Expired due to failure to pay maintenance fee|
Effective date: 19920131